A few days ago, I posted a summary of some recent work on syntactic alignment with Kodi Weatherholtz and Kathryn Campell-Kibler (both at The Ohio State University), in which we used the WAMI interface to collect speech data for research on language production over Amazon’s Mechanical Turk.
At AMLAP 2013 in Marseilles I also presented joint research with Jane Grimshaw (at Rutgers University) that employed a similar over-the-web paradigm to record spoken sentence productions as part of a sentence recall task. We investigated that-mention in complement clauses as a function of the embedding-verb. Unlike in previous psycholinguistic studies on this topic, we employed a large number of complement clause-embedding verbs (64 different verbs, spanning two register and several semantic classes). We find effects of the verb’s semantic class and register. We also find effects of the complement clause’s predictability given the embedding verb as well as the frequency of the verb. We analyze both that-omission and that–omissibility (whether a verb allows omission at all). While we’re still awaiting the final analysis, the preliminary findings largely show the same results for both variables. I’m posting this now simply as another example of a paradigm that incorporates the recording of speech over the web into a classic psycholinguistic paradigm.
Comments welcome. We’re still writing up the results.